All video posts

Critical thinking · Junior developers · AI

Junior Hires Can't Think on Their Own Any More. Here's the Fix

A lot of junior hires who studied and graduated in the ChatGPT era are not really thinking any more, and the scary part is that they do not know it. We have optimised so hard for speed and output that we are outsourcing our brains. Thinking is a skill, and it can be trained back. Here is how I do it.

Deric YeeDeric Yee 16 November 2026 6 min read 8 min video

Prefer to read? The full article is below. Watch on YouTube.

The short version

  1. 1Many juniors look for answers instead of reasoning towards them, and AI lets them skip the thinking entirely.
  2. 2If everyone uses the same templates and hacks, none of them differentiate you. Speed without thought is noise.
  3. 3A good hire asks what to do. A great hire asks why, so they can suggest a better how.
  4. 4Test thinking in interviews with vague, open-ended problems and watch for clarifying questions.
  5. 5Train it with first principles, "why, what if, then what" loops, shadow-and-debrief, and SOPs that always explain the reasoning.

The death of thinking in junior hires

I have hired, trained and worked with a lot of juniors and seniors across my businesses, and I keep seeing the same pattern. I once asked a junior team member to evaluate two AI tools and pick the better one for writing cold emails. They came back with: "Tool A has more features, Tool B is more popular, so I choose B", plus a ChatGPT summary of both. No testing, no metrics, no cost-benefit analysis, no first-principles reasoning.

This is happening everywhere. Juniors are trained to look for answers, not to create them. Schools taught memorisation more than problem solving, and now tools like ChatGPT let people skip the thinking process altogether.

I am all for speed. But if someone cannot explain why they are doing what they are doing, they are not growing. They are copy-pasting themselves into mediocrity. If you do not catch it early, your whole team becomes a glorified assembly line of prompters who ship fast and break things along the way. This is the same risk I describe in why you should not become dependent on AI.

Copying without a hypothesis

I once walked into a product sprint meeting with our developers, product leads and a designer. They were deciding where to place a feature based on what a competitor had done. Not user behaviour, not data. Vibes and screenshots of someone else's app.

I asked why we were copying them again. My motto at the company is literally innovate or die. If someone is doing something well, copy it, absolutely. But understand it and have your own hypothesis or first-principles reason for why you are copying it. "They do it, so we should" is not a reason.

That was when I realised thinking is not automatic. You have to train it.

“Speed without thought is just noise.”

Deric Yee, in the video

The danger of over-optimisation

We live in a world of templates, tricks and hacks: cheat sheets, prompt libraries, "top 10 growth strategies". But if everyone uses the same tactics, none of them can differentiate you. Optimisation without understanding is just laziness in disguise. You cannot stand out by copying.

We have built systems that reward action over intention. So in my team, we keep asking: why are we doing this? We use Loom to record short, roughly three-minute breakdowns explaining the thinking behind decisions, so everyone understands not just the what and the how, but the why.

That habit forces clarity and kills shallow thinking. It also makes mistakes visible early. When you can see someone's reasoning, you can spot the wrong assumption behind it. Without the reasoning, you trust gut feel and only discover the problem once it has already blown up.

Hire for the why

Most companies hire as if they are assembling IKEA furniture: here is the SOP, follow it. Across sales, operations, development and product, I have learned that a good hire asks what to do, but a great hire asks why. Only by understanding the why can they offer the best view on how, because they are the ones doing the work and often know a better way. That difference decides whether someone grows with your company or stalls it.

In interviews I give a vague, open-ended problem, for example: "I am launching a new course and you have a $10,000 budget. What is your plan?" Then I watch them think. Do they ask clarifying questions? Do they break it down from first principles: audience, cost per lead, clarity of the offer? If they jump straight to tactics like "run Facebook ads" or "hire an agency", that is a red flag.

This is not just leadership talk. McKinsey research has linked a culture of critical thinking with stronger innovation and growth. If you are a junior preparing for interviews, practise exactly this kind of reasoning. Our guide to what hiring managers look for goes deeper.

A system to train thinking back

This is the system I actively use in my company. We have team members from several countries with different upbringings and education, so some people find first-principles thinking harder at first. That is fine. It can be taught, and as a leader it is your job to teach it and to keep improving at it yourself.

01

Teach first-principles thinking

Break problems to the root. We are not "trying to get more leads"; we are matching a message to a market that wants it enough to give us time or money. That reframe opens new options. Books on mental models, such as Farnam Street's The Great Mental Models, help build the muscle.

02

Use thinking loops

A simple tool: why are we doing this, what if we do not, and then what happens? It forces people to forecast consequences and weigh decisions.

03

Shadow and debrief

A junior shadows a senior, then explains back not just what happened but why they think it was done that way. That builds metacognition: thinking about thinking.

04

Never write an SOP without context

Every SOP in my company comes with a one- to two-minute Loom explaining the reasoning. If you want robots, use robots. With humans, you want judgement.

AI cannot take responsibility

ChatGPT can write emails and generate ideas. What it cannot do is take responsibility when a strategy burns $10,000 on ads and nobody signs up. That is on a person. You need people you can delegate to and hold accountable, which means thinkers, not just prompt engineers.

In a world where anyone can press a button and get an answer, asking the right question is the superpower. Ask the wrong one and you are rowing hard in the wrong direction. For developers in particular, being able to explain the code you built with AI is the clearest evidence that you actually thought it through.

Let thinking make a comeback

To recap: many juniors today do rather than think. Over-optimisation kills curiosity and creativity. Hire people who reason from principles instead of copy-pasting strategies, and build systems that reward clarity, not just speed.

Next time you hand someone a task, ask yourself: can this person explain why it matters? If not, there is work to do. Thinking is the most expensive skill because it takes time, effort and humility. It is also the most valuable.

If you are early in your career and want to build that thinking muscle alongside technical skills, that is how we teach at Sigmaschool: every project asks you to explain your reasoning, not just ship output. You can experience it with the free trial.

Questions people ask

  • Why do junior employees struggle with critical thinking?

    Many were taught to find correct answers rather than reason through open problems, and AI tools now make it easy to skip the reasoning step entirely. The result is work that looks finished but lacks testing, metrics or a clear rationale. It is not a fixed trait. Critical thinking is a skill that improves quickly with the right habits and feedback.

  • How can I test critical thinking in an interview?

    Give candidates a vague, open-ended scenario, such as planning a product launch with a fixed budget, and watch how they approach it. Strong thinkers ask clarifying questions and break the problem into fundamentals like audience, costs and the offer before suggesting tactics. Jumping straight to a generic tactic without understanding the problem is a warning sign.

  • What is first-principles thinking?

    First-principles thinking means breaking a problem down to its most basic truths and reasoning up from there, instead of copying what others do. For example, instead of "we need more leads", you ask what actually produces a lead: a message that matches a market that wants it. Reframing problems this way often reveals options that copying would miss.

  • Does using AI make people worse at thinking?

    It can, if people use it to skip reasoning rather than support it. Accepting AI output without testing it, questioning it or understanding why it works weakens your judgement over time. Used well, AI can speed up research and drafting while you stay responsible for the decision. The key is always being able to explain why.

Your next step

Our 12-week AI-Native Software Development Programme is a live, mentor-reviewed cohort where you build and ship real products with AI, and learn to explain every line. Start with the free trial to see how we teach, then decide.

Watch next

All videos

Learn the fundamentals that pay off in any direction.
Coding plus AI, taught properly.

Whatever you end up doing, the people who understand how software and AI actually work get to direct them. Try the first lessons free, no card needed, or see the full 12-week AI-Native Software Development Programme.